[09:14:38] <grrrit-wm>	 (CR) Nuria: [C: 2] "Very good catch. Merging." [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149383 (owner: QChris)
[09:14:49] <grrrit-wm>	 (Merged) jenkins-bot: Update the database connection settings for alembic [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149383 (owner: QChris)
[09:20:21] <grrrit-wm>	 (CR) Nuria: "On my opinion to merge performance fixes we need to do more than one trivial test. More so with the OOMs we have seen in staging as of yes" [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149482 (https://bugzilla.wikimedia.org/68596) (owner: Milimetric)
[09:30:17] <grrrit-wm>	 (CR) QChris: "> On my opinion to merge performance fixes we need to do more than one trivial test." [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149482 (https://bugzilla.wikimedia.org/68596) (owner: Milimetric)
[09:32:37] <grrrit-wm>	 (CR) Nuria: [C: 2] "Thanks for documenting this." [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149298 (owner: QChris)
[09:32:46] <grrrit-wm>	 (Merged) jenkins-bot: Document choice of constant in connection pool test [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149298 (owner: QChris)
[09:34:48] <qchris>	 nuria: Since we cannot move repos in gerrit ... Are we settled for the name dashiki?
[09:34:59] <qchris>	 I thought milimetric does not like the name to much.
[09:35:07] <qchris>	 (But maybe that was a misunderstanding on my part)
[09:35:16] <nuria>	 qchris: it was me who did not love it
[09:35:24] <nuria>	 but i am happy to go along
[09:35:35] <qchris>	 Mhmm. ok then.
[09:35:38] <nuria>	 dan choose it
[09:35:45] <nuria>	 and kevin said (gimme a sec)
[09:36:18] <nuria>	 "BTW I love the name dashiki"
[09:36:31] <nuria>	 but wait ....
[09:36:44] <nuria>	 what dan said was that he liked
[09:36:50] <nuria>	 qchris: but it was taken
[09:36:58] <nuria>	 that is what it was, my mistake
[09:37:25] <nuria>	 dan said " it's taken already: https://github.com/brewster/dashiki  (not too active, I think we can steal it, but still)"
[09:37:50] <nuria>	 so i guess we better wait
[09:38:02] <qchris>	 Yes. That does not sound like a "We want dashiki as name"
[09:38:14] <qchris>	 But too late: https://gerrit.wikimedia.org/r/#/admin/projects/analytics/dashiki
[09:38:28] <nuria>	 ok -> nuria takes full blame
[09:38:53] <qchris>	 We can remove it later if we have to, and create it under a new name.
[09:38:53] <nuria>	 and apologizes
[09:39:00] <qchris>	 No worries.
[09:39:18] <nuria>	 let's leave it on 'stand-by' and see what dan and kevin think
[09:59:58] <grrrit-wm>	 (CR) Nuria: [C: 2] Drop unneeded imports in connection pool test (1 comment) [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149299 (owner: QChris)
[10:00:16] <grrrit-wm>	 (Merged) jenkins-bot: Drop unneeded imports in connection pool test [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149299 (owner: QChris)
[10:11:38] <grrrit-wm>	 (CR) Nuria: [C: 2] Be lighter on database when testing connection pools [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149300 (owner: QChris)
[10:11:47] <grrrit-wm>	 (Merged) jenkins-bot: Be lighter on database when testing connection pools [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149300 (owner: QChris)
[12:45:54] <wikibugs>	 Analytics / Quarry: Provide the ability to view the queries submitted by a particular user - https://bugzilla.wikimedia.org/69174 (Aaron Halfaker) NEW p:Unprio s:normal a:None See summary.
[12:46:07] <wikibugs>	 Analytics / Quarry: Provide the ability to view the queries submitted by a particular user - https://bugzilla.wikimedia.org/69174 (Aaron Halfaker) s:normal>enhanc
[12:50:36] <grrrit-wm>	 (PS1) QChris: Reindent oozie webrequest dataset definiton [analytics/refinery] - https://gerrit.wikimedia.org/r/152048
[12:50:38] <grrrit-wm>	 (PS1) QChris: Add Oozie webrequest dataset with done marker [analytics/refinery] - https://gerrit.wikimedia.org/r/152049
[12:50:40] <grrrit-wm>	 (PS1) QChris: [WIP] Notify Icinga about done webrequest datasets [analytics/refinery] - https://gerrit.wikimedia.org/r/152050
[12:53:34] <grrrit-wm>	 (CR) QChris: "> I think it would better to have icinga monitoring separate. > Outside of oozie/webrequest/partition/add." (1 comment) [analytics/refinery] - https://gerrit.wikimedia.org/r/151957 (owner: Ottomata)
[12:55:22] <grrrit-wm>	 (CR) QChris: [WIP] Notify Icinga about done webrequest datasets (1 comment) [analytics/refinery] - https://gerrit.wikimedia.org/r/152050 (owner: QChris)
[12:59:42] <grrrit-wm>	 (PS1) QChris: Drop instructions for manual port forwarding in Vagrant [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/152057
[13:00:23] <grrrit-wm>	 (PS2) QChris: Drop instructions for manual port forwarding in Vagrant [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/152057
[13:11:25] <wikibugs>	 Analytics / Quarry: "Running" query being displayed as unsubmitted - https://bugzilla.wikimedia.org/69176 (TheOriginalSoni) UNCO p:Unprio s:normal a:None When I try to access a "running" query through http://quarry.wmflabs.org/query/runs/all the query it does not display as running but as a...
[13:14:56] <grrrit-wm>	 (PS1) Yuvipanda: Add new runner node to fabfile [analytics/quarry/web] - https://gerrit.wikimedia.org/r/152060
[13:14:58] <grrrit-wm>	 (PS1) Yuvipanda: Fix missed rename of config.py to config.yaml [analytics/quarry/web] - https://gerrit.wikimedia.org/r/152061
[13:15:33] <grrrit-wm>	 (CR) Yuvipanda: [C: 2] Add new runner node to fabfile [analytics/quarry/web] - https://gerrit.wikimedia.org/r/152060 (owner: Yuvipanda)
[13:15:38] <grrrit-wm>	 (Merged) jenkins-bot: Add new runner node to fabfile [analytics/quarry/web] - https://gerrit.wikimedia.org/r/152060 (owner: Yuvipanda)
[13:15:42] <grrrit-wm>	 (CR) Yuvipanda: [C: 2] Fix missed rename of config.py to config.yaml [analytics/quarry/web] - https://gerrit.wikimedia.org/r/152061 (owner: Yuvipanda)
[13:15:46] <grrrit-wm>	 (Merged) jenkins-bot: Fix missed rename of config.py to config.yaml [analytics/quarry/web] - https://gerrit.wikimedia.org/r/152061 (owner: Yuvipanda)
[13:23:42] <grrrit-wm>	 (CR) Ottomata: [C: 2 V: 2] Reindent oozie webrequest dataset definiton [analytics/refinery] - https://gerrit.wikimedia.org/r/152048 (owner: QChris)
[13:26:59] <grrrit-wm>	 (CR) Ottomata: "Good idea!" (2 comments) [analytics/refinery] - https://gerrit.wikimedia.org/r/152049 (owner: QChris)
[13:30:36] <ottomata>	 yoo qchris, mornin!
[13:30:46] <qchris>	 Just startet "/msg ottomata" :-)
[13:30:52] <qchris>	 Good morning.
[13:31:04] <qchris>	 You beat me.
[13:31:15] <ottomata>	 i like your changes!
[13:31:27] <ottomata>	 i don't like the fact that I will now have to manage not 4 but 8 different coordinators
[13:31:30] <qchris>	 I am not completely sure about the failures and Icinga :-/
[13:31:32] <ottomata>	 but your bundles are so good that it isn't even a thang
[13:31:36] <ottomata>	 naw, i think that's fine
[13:31:48] <ottomata>	 that's the whole point of using the refresh interval passive check
[13:31:56] <ottomata>	 this is a pretty crude alert anyway
[13:31:59] <qchris>	 Oh I guess we'll get even more coordinators over time :-)
[13:32:08] <ottomata>	 all it really does is tell us to look at what's wrong
[13:32:15] <ottomata>	 so, if it gets stale and we get an alert
[13:32:16] <ottomata>	 so be it!
[13:32:20] <ottomata>	 so, idea!
[13:32:23] <wikibugs>	 Analytics / Tech community metrics: Remove severity related graphs from bugzilla_response_time.html - https://bugzilla.wikimedia.org/69179 (Andre Klapper) p:Unprio>Low
[13:32:24] <wikibugs>	 Analytics / Tech community metrics: Remove severity related graphs from bugzilla_response_time.html - https://bugzilla.wikimedia.org/69179 (Andre Klapper) NEW p:Unprio s:enhanc a:None Having used http://korma.wmflabs.org/browser/bugzilla_response_time.html for a while, Quim and I have reali...
[13:32:28] <ottomata>	 can we make the monitor done flag thing more generic
[13:32:30] <grrrit-wm>	 (CR) Nuria: [C: 2] Drop noop assertion in connection pool test [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149301 (owner: QChris)
[13:32:33] <ottomata>	 and not webrequest specifc?
[13:32:52] <qchris>	 We could. But wouldn't that be premature optimization?
[13:32:55] <grrrit-wm>	 (Merged) jenkins-bot: Drop noop assertion in connection pool test [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149301 (owner: QChris)
[13:32:59] <ottomata>	 haha
[13:33:00] <ottomata>	 maybe so.
[13:33:04] <ottomata>	 the bundle would be webrequest specific
[13:33:17] <qchris>	 And datasets too.
[13:33:29] <ottomata>	 right, but that would be provided by the properites file, yes?
[13:33:38] <ottomata>	 $datasets_file set in bundle.properties
[13:33:41] <qchris>	 Yes.
[13:33:41] <ottomata>	 only q
[13:33:46] <ottomata>	 can this
[13:33:50] <ottomata>	         <data-in name="input" dataset="webrequest">
[13:33:51] <ottomata>	 be variable?
[13:34:00] <qchris>	 But if we make it more generic, we'd have to pass the right datasets file to the right coordinators within the bundel.
[13:34:02] <ottomata>	 $dataset_name
[13:34:16] <qchris>	 I think so, but I didn't try.
[13:34:17] <ottomata>	 ja bundle.xml is webrequest too
[13:34:35] <qchris>	 We already use EL in attribute values, so it should work.
[13:35:04] <ottomata>	 the $webrequest_source param in workflow.xml coul djust be '$icinga_alert_name' or description, or something
[13:35:16] <ottomata>	 qchris: you have labs instances up where you test this?
[13:35:40] <qchris>	 qchris-master + qchris-worker{1,2,3,4,5,6}.
[13:35:46] <qchris>	 It is actually tested.
[13:36:17] <ottomata>	 ok awesome, mind if I try this there then?
[13:36:50] <qchris>	 Not sure what you mean ... but run wild on this test cluster.
[13:36:53] <ottomata>	 ha, ok
[13:37:00] <ottomata>	 i mean, trying to make this generic, and testing it out there
[13:37:04] <ottomata>	 oh, first, real quick
[13:37:07] <ottomata>	 let's bikeshed this
[13:37:07] <ottomata>	 https://gerrit.wikimedia.org/r/#/c/152049/1
[13:37:19] <ottomata>	 https://gerrit.wikimedia.org/r/#/c/152049/1/oozie/webrequest/datasets.xml
[13:37:37] <qchris>	 I do not like "making it more generic" at this point too much. Let's postpone making it even more generic until we need it.
[13:37:54] <qchris>	 webrequest_unchecked, and webrequest_raw sound good to me.
[13:38:16] <qchris>	 Meh.  webrequest_raw would probably
[13:38:21] <qchris>	 get confused with the raw table.
[13:38:33] <qchris>	 So what about your first suggestion "webrequest_unchecked"?
[13:39:16] <qchris>	 Or mayby even "webrequest_raw_unchecked" and the checked one "webrequest_raw�?
[13:39:21] <ottomata>	 hm, maybe
[13:39:28] <ottomata>	 might make sense in the future
[13:39:45] <ottomata>	 actually, i'm ok with webrequest_no_done_flag the more i think about it...
[13:39:46] <ottomata>	 hmmm
[13:39:47] <ottomata>	 dunno
[13:39:55] <qchris>	 :-D
[13:39:59] <ottomata>	 webrequest_unverified?
[13:40:00] <ottomata>	 hm
[13:40:48] <qchris>	 "..._unverified" is ok with me to.
[13:41:18] <qchris>	 "..._unchecked" sounds a bit better, but
[13:41:33] <ottomata>	 unchecked
[13:41:34] <ottomata>	 hm
[13:41:42] <qchris>	 The datasets would continue to be datasets in "..._unchecked" even after we checked them.
[13:41:54] <qchris>	 Then again ... the same also holds true for "..._unverified"
[13:42:14] <ottomata>	 yes, but verified somehow sounds more official than checked
[13:42:24] <qchris>	 Then we should use checked :-)
[13:42:32] <ottomata>	 especially re. your point about how 0.0 does not necessarily mean 100% good data :p
[13:42:35] <qchris>	 It's just some automatic duplicate check.
[13:42:38] <ottomata>	 could mean loss + duplicates
[13:42:52] <qchris>	 No real verification. Right.
[13:43:13] <ottomata>	 ok, unchecked
[13:43:34] <qchris>	 and is it ok if I additionally rename "webrequest" to "webrequest_raw" for the datasets?
[13:43:41] <qchris>	 Cool. unchecked it is.
[13:44:13] <ottomata>	 hmmmm
[13:44:19] <ottomata>	 not sure, i'm ok with it i guess
[13:44:26] <ottomata>	 might be nice if/when we have a real webrequest non-raw table later
[13:44:35] <qchris>	 If you're not sure, we'll leave things as they are for now.
[13:44:40] <qchris>	 We can always do it later.
[13:44:52] <ottomata>	 ok
[13:45:01] <ottomata>	 so, no generic making, eh?!
[13:45:04] <ottomata>	 dawwwww
[13:45:08] <ottomata>	 but it looks so close here
[13:45:12] <qchris>	 We are already sooooo generic.
[13:45:24] <qchris>	 I wanted to bikeshed that a bit more, but we ran out of time.
[13:45:40] <qchris>	 Having things like workflows as parameters to the coordinators and such ...
[13:45:56] <qchris>	 It does not help me at all when testing.
[13:46:44] <qchris>	 So at some point in time, I'll gonna start discussing those and making our property files more compact again :-)
[13:47:24] <ottomata>	 the alert in icinga is generic!
[13:47:24] <ottomata>	 https://gerrit.wikimedia.org/r/#/c/151963/1/templates/icinga/checkcommands.cfg.erb
[13:47:31] <ottomata>	 wait, not sure what you mean,
[13:47:51] <ottomata>	 oh, hm, that stuff helped me, because I was able to use the same files for testing in production cluster
[13:47:55] <ottomata>	 but under my user and my own app paths
[13:48:28] <qchris>	 I am doing the same thing ... I just override oozie_directory and I am done with.
[13:48:51] <ottomata>	 ja i suppose
[13:48:52] <ottomata>	 ok ok
[13:48:53] <qchris>	 I do not override workflow_file and such.
[13:48:58] <ottomata>	 let's bike shed that later, as you say
[13:49:02] <qchris>	 Right.
[13:49:02] <grrrit-wm>	 (CR) Nuria: Bring connection pool tests into main test suite (1 comment) [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149302 (owner: QChris)
[13:49:06] <qchris>	 :-)
[13:49:21] <ottomata>	 but, you think I should not try to make this icinga check generic, eh?
[13:49:35] <ottomata>	 its so easy though!  just have the coordinators and workflows look for _SUCCESS file in any dataset
[13:49:38] <qchris>	 Your call. Or ops call rather :-)
[13:49:47] <ottomata>	 and send alert name based on property to icinga
[13:49:49] <ottomata>	 haha
[13:49:52] <qchris>	 I didn't push back that the other script even had "--not-exists"
[13:50:21] <qchris>	 Wait ... daily scrum is in 10 minutes ... gonna grab something to eat
[14:00:53] <nuria>	 sigining again kevinator as i cannot join standup
[14:01:59] <terrrydactyl>	 also says i'm not allowed to join hangout
[14:02:17] <kevinator>	 I am having difficulty as well
[14:02:24] <wikibugs>	 Analytics / Wikimetrics: tox runs all tests (including manual ones) - https://bugzilla.wikimedia.org/69183 (nuria) NEW p:Unprio s:normal a:None tox runs all tests (including manual ones).  We should make sure that when running tox the tests on the manual directory are not run, behaviour sho...
[14:03:19] <nuria>	 close browser, sign out and sign in
[14:03:23] <nuria>	 kevinator
[14:03:32] <nuria>	 and terrydactyl
[14:03:38] <nuria>	 terrrydactyl
[14:05:11] <qchris_meeting>	 terrrydactyl: I reinvited you. Can you try again.
[14:31:45] <grrrit-wm>	 (CR) Nuria: [C: 2] "Leaving aside the fact that tox runs all tests when it shouldn't this looks good." (1 comment) [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149384 (owner: QChris)
[14:31:54] <grrrit-wm>	 (Merged) jenkins-bot: Allow to run nosetests without coverage report [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149384 (owner: QChris)
[14:38:06] <grrrit-wm>	 (PS2) Ottomata: Add Oozie webrequest dataset with done marker [analytics/refinery] - https://gerrit.wikimedia.org/r/152049 (owner: QChris)
[14:43:08] <grrrit-wm>	 (CR) QChris: [C: 2 V: 2] "Thanks!" [analytics/refinery] - https://gerrit.wikimedia.org/r/152049 (owner: QChris)
[14:43:43] <grrrit-wm>	 (CR) Nuria: "Ok , let me phrased it differently: I do not think is enough to test 1 metric in 1 cohort. I think we should test the metric improvements " [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149482 (https://bugzilla.wikimedia.org/68596) (owner: Milimetric)
[14:45:55] <ottomata>	 ah!
[14:45:57] <ottomata>	 i just added comments!
[14:45:58] <ottomata>	 :p
[14:46:00] <grrrit-wm>	 (PS2) QChris: [WIP] Notify Icinga about done webrequest datasets [analytics/refinery] - https://gerrit.wikimedia.org/r/152050
[14:46:27] <qchris>	 Wasn't that meant to be merged?
[14:46:44] <qchris>	 You uploaded a patch set with changes we talked about.
[14:46:55] <qchris>	 I was preparing it too ... but you beat me :-)
[14:48:03] <grrrit-wm>	 (PS1) Ottomata: Add documentation for webrequest datasets [analytics/refinery] - https://gerrit.wikimedia.org/r/152094
[14:48:05] <ottomata>	 haha
[14:48:17] <qchris>	 I guess ... Don't upload new patch sets on changes I started. I am a ruthless merger.
[14:49:07] <grrrit-wm>	 (CR) QChris: [C: 2 V: 2] Add documentation for webrequest datasets [analytics/refinery] - https://gerrit.wikimedia.org/r/152094 (owner: Ottomata)
[14:50:49] <grrrit-wm>	 (CR) QChris: Add Oozie webrequest dataset with done marker (2 comments) [analytics/refinery] - https://gerrit.wikimedia.org/r/152049 (owner: QChris)
[14:54:38] <grrrit-wm>	 (CR) Nuria: "I will investigate using celery groups rather than chains as it seems a better fit for what we want to do." (1 comment) [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/150475 (https://bugzilla.wikimedia.org/68840) (owner: Milimetric)
[14:55:37] <nuria>	 qchris: how can we ensure teresa has permits to push code to operations/puppet/wikimetrics like we do?
[14:56:23] <qchris>	 By "push" you mean "Being able to CR+2", or you mean "upload now commits"?
[14:57:00] <qchris>	 s/now/new/
[14:57:17] <ottomata>	 haha
[14:57:17] <terrrydactyl>	 upload new commits
[14:57:30] <terrrydactyl>	 i'm trying to show nuria what i've changed to puppet
[14:57:33] <ottomata>	 i think anyone can upload changes
[14:57:36] <terrrydactyl>	 but it won't let me push any of my code
[14:57:38] <YuviPanda>	 w00t, quarry is being used nicely! Number of new user registrations per day since the beginning! http://quarry.wmflabs.org/query/100
[14:57:43] <ottomata>	 terrrydactyl:
[14:57:45] <qchris>	 I thought so too.
[14:57:51] <ottomata>	 git add remote gerrit ssh://blablabla
[14:57:54] <ottomata>	 whatever your ssh url is
[14:57:55] <ottomata>	 for that repo
[14:58:00] <ottomata>	 with your username in it
[14:58:22] <ottomata>	 https://gerrit.wikimedia.org/r/#/admin/projects/operations/puppet/wikimetrics
[14:58:25] <ottomata>	 you an see it there
[14:58:33] <ottomata>	 then you should be able to use git review like usual
[14:58:35] <ottomata>	 (if you use that)
[14:59:03] <ottomata>	 (that is while cd-ed into puppet/modules/wikimetrics)
[15:00:05] <qchris>	 YuviPanda: \o/ Party!
[15:00:10] <YuviPanda>	 qchris: :D \o/
[15:03:04] <grrrit-wm>	 (PS3) Ottomata: [WIP] Notify Icinga about done webrequest datasets [analytics/refinery] - https://gerrit.wikimedia.org/r/152050 (owner: QChris)
[15:03:34] <terrrydactyl>	 ottomata: tried that and got `fatal: pathspec 'remote' did not match any files`
[15:04:01] <qchris>	 "git add remote [...]" should be "git remote add [...]"
[15:04:14] <leila>	 nuria, re your email about android vs ios
[15:04:19] <ottomata>	 whoops
[15:04:25] <ottomata>	 sorry thanks qchris
[15:04:47] <ottomata>	 terrrydactyl: ^ do as qchris says :)
[15:04:52] <terrrydactyl>	 :)
[15:04:53] <leila>	 how do you want to proceed? we can recompute this number based on Oliver's latest data
[15:05:05] <qchris>	 :-P Do not listen to qchris.
[15:05:39] <terrrydactyl>	 i got this now: http://paste.ofcode.org/jHLMBa8cDPMtSgdurVyBy4
[15:07:25] <nuria>	 terrrydactyl i do not use git review
[15:07:47] <nuria>	 but can't you not push directly like git push origin HEAD:refs/for/master
[15:07:47] <yurikR>	 noone ever listens to qchris
[15:07:53] <qchris>	 terrrydactyl: Looks like you didn't use the proper url when adding the remote.
[15:08:24] <qchris>	 terrrydactyl: Did you use the ssh url from the project page?
[15:08:47] <nuria>	 you can use 'git remote show origin' and see what you have
[15:08:53] <terrrydactyl>	 qchris, from this page? https://gerrit.wikimedia.org/r/#/admin/projects/operations/puppet/wikimetrics
[15:09:16] <ottomata>	 yes, your url should be
[15:09:17] <qchris>	 Yes. The url you used should look like ssh://terrrydactyl@gerrit.wikimedia.org:29418/operations/puppet/wikimetrics
[15:09:27] <ottomata>	 ja, (if terrrydactyl is your gerrit username)
[15:09:31] <qchris>	 (If terrrydactyl is your username)
[15:09:33] <qchris>	 Right.
[15:09:34] <ottomata>	 hah
[15:10:40] <terrrydactyl>	 i added as `wikimetrics git:teresa-puppet-changes ❯ git remote add gerrit ssh://terrrydactyl@gerrit.wikimedia.org:29418/operations/puppet/wikimetrics` so it should be right..
[15:11:00] <ottomata>	 hm
[15:11:27] <terrrydactyl>	 oh!
[15:11:30] <terrrydactyl>	 no, i didn't
[15:11:32] <qchris>	 The paste your gave us, says "Project not found: r/operations/puppet/wikimetrics" (note the leading "r/")
[15:11:41] <terrrydactyl>	 how do i reset the remote?
[15:12:32] <qchris>	 git remote remove gerrit
[15:12:42] <qchris>	 terrrydactyl: ^
[15:13:16] <ottomata>	 qchris, re the icinga oozie stuff
[15:13:19] <ottomata>	 in the interest of simplicity
[15:13:25] <ottomata>	 why not just do something like
[15:13:53] <ottomata>	 end_shell_action_script = 'send_icinga.sh'
[15:13:53] <ottomata>	 end_shell_action_arg = $webrequest_source
[15:13:59] <ottomata>	 on the main bundle/ workflow
[15:14:43] <ottomata>	 and then just have a generic shell action step if mark_dataset_done works
[15:15:05] <qchris>	 That's way too generic for me, and buys us 2 parameters.
[15:15:26] <ottomata>	 its generic, but very simple
[15:15:32] <qchris>	 Those parameters have to be added in bundle, coordinator, and workflow xmls.
[15:15:34] <ottomata>	 you can just pass /bin/true if you don't want to do anything
[15:15:39] <ottomata>	 jhm
[15:15:39] <ottomata>	 hm
[15:15:52] <qchris>	 Also /bin/true won't work.
[15:16:04] <qchris>	 Paths always failed for me.
[15:16:15] <ottomata>	 oh, the script has to be in hdfs with the app path
[15:16:16] <ottomata>	 ?
[15:16:16] <qchris>	  /bin/echo did not work for example.
[15:16:21] <qchris>	 Yes.
[15:16:35] <qchris>	 cwd on the node.
[15:16:41] <terrrydactyl>	 ottomata, qchris. huzzah, it worked!
[15:16:44] <terrrydactyl>	 thanks :)
[15:16:46] <ottomata>	 oook fine, but I'm going to try to make the separate bundle/workflow generic then!
[15:16:47] <qchris>	 terrrydactyl: yay!
[15:16:49] <ottomata>	 just going to try....:)
[15:17:15] <qchris>	 What's wrong with hardcoding it for now?
[15:17:23] <qchris>	 We do not need it parametrized for now, do we?
[15:18:09] <ottomata>	 well, ok, i think one of the benifits of designing things generically from the start is that you don't break them when you have to do it later
[15:18:18] <ottomata>	 i'm with you for many things, we don't want to overengineer
[15:18:25] <ottomata>	 but in this case, it will not be harder to make this generic
[15:18:36] <ottomata>	 and this will likely be very useful later
[15:18:43] <ottomata>	 like, when we want to monitor the success of the refined dataset
[15:19:10] <qchris>	 You want to use this for debugging. Right? Why not make a debug parameter, that the send_ok_to_icinga.sh accepts?
[15:19:22] <qchris>	 That's less parameters, and still solves the problem.
[15:19:48] <ottomata>	 eh?  i'm just talking about making the monitor_done_flag workflow non-webrequest specific, that's all
[15:20:15] <ottomata>	 all it does it check for a _SUCCESS flag in a dataset
[15:20:19] <ottomata>	 that is a pretty generic use case
[15:20:51] <qchris>	 Ok. As you say. I guess I'll be convinced by your code.
[15:20:59] <qchris>	 :-)
[15:28:05] <grrrit-wm>	 (CR) Nuria: [WIP] Add ability to global query a user's wikis (3 comments) [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/129858 (owner: Terrrydactyl)
[15:32:39] <wikibugs>	 Analytics / Quarry: List all queries - https://bugzilla.wikimedia.org/69189 (TheOriginalSoni) UNCO p:Unprio s:normal a:None When I use http://quarry.wmflabs.org/query/runs/all it only shows me the last 25 queries. Is there any way to see all queries?
[15:37:57] <grrrit-wm>	 (PS3) QChris: Put focus on last connection in connection pool test [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149303
[15:37:59] <grrrit-wm>	 (PS2) QChris: Bring connection pool tests into main test suite [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149302
[15:39:30] <grrrit-wm>	 (CR) QChris: Bring connection pool tests into main test suite (1 comment) [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149302 (owner: QChris)
[15:40:12] <ottomata>	 qchris, do the commands used inside of the shell script need to be in hdfs too?
[15:40:14] <ottomata>	 no, right?
[15:40:23] <qchris>	 They have to :-(
[15:40:28] <qchris>	 Well ...
[15:40:29] <ottomata>	 send_nsca?
[15:40:31] <ottomata>	 hostname?
[15:40:35] <ottomata>	  echo?
[15:40:41] <qchris>	 If they are on the path, that's fine.
[15:40:48] <ottomata>	 ok
[15:41:27] <qchris>	 But if they are in hdfs, the action canbe made to copy it to the node to the right spot and run it from there. So having it in hdfs has quite some benefits.
[15:42:19] <qchris>	 But you're right. send_nsca in hdfs makes no sense.
[15:42:56] <qchris>	 ottomata: Sorry. Misread your questions. /Inside the shell script/, you can do whatever you want.
[15:43:01] <ottomata>	 you sure?
[15:43:04] <qchris>	 Yes.
[15:43:08] <ottomata>	 i can reference absolute paths
[15:43:11] <qchris>	 Yes.
[15:43:14] <ottomata>	 oh hm, and this is going to be run on any node, eh?
[15:43:16] <ottomata>	 not just oozie server
[15:43:22] <qchris>	 Any node.
[15:43:24] <ottomata>	 so i need to make sure send_nsca is installed everywhere
[15:43:29] <ottomata>	 echo and hostname we can be confident about
[15:43:54] <qchris>	 I tried installing the bundle on a 1 master + 6 slaves node. And the shell script was run locally on 4 different nodes for 5 action invocations.
[15:44:00] <ottomata>	 and, i need to hardcode the hostname that we use in icinga
[15:44:05] <qchris>	 Right.
[15:44:05] <ottomata>	 so that it always uses the same hostname
[15:44:07] <ottomata>	 ok
[15:48:13] <grrrit-wm>	 (PS9) Terrrydactyl: [WIP] Add ability to global query a user's wikis [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/129858
[15:51:52] <terrrydactyl>	 nuria ^
[15:52:27] <nuria>	 yes
[15:52:55] <nuria>	 terrrydactyl did you corrected all session issues and things work well now?
[15:53:22] <terrrydactyl>	 things seem to be working with the if statement fine
[15:53:52] <grrrit-wm>	 (CR) Nuria: [C: 2] "thanks for the fast turnaround." [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149302 (owner: QChris)
[15:54:00] <grrrit-wm>	 (Merged) jenkins-bot: Bring connection pool tests into main test suite [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/149302 (owner: QChris)
[15:56:55] <nuria>	 ok terrrydactyl will look at changes
[16:06:48] <grrrit-wm>	 (CR) Nuria: [WIP] Add ability to global query a user's wikis (4 comments) [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/129858 (owner: Terrrydactyl)
[16:10:57] <nuria>	 qchris: regarding the Vagrantfile changes
[16:10:59] <nuria>	 https://gerrit.wikimedia.org/r/#/c/150487/2/puppet/manifests/roles/wikimetrics.pp
[16:11:12] <nuria>	 we did test that those do not break on staging/dev?
[16:11:13] <grrrit-wm>	 (PS4) Ottomata: [WIP] Notify Icinga about done webrequest datasets [analytics/refinery] - https://gerrit.wikimedia.org/r/152050 (owner: QChris)
[16:11:28] <nuria>	 sorry qchris
[16:37:52] <wikibugs>	 Analytics / Wikimetrics: session management - https://bugzilla.wikimedia.org/68833#c3 (nuria) Note that flask-sqlalchemy will help us with sessions instantiated by a web request. But not with sessions instantiated by the queue which does not interact with flask.  I think that tasks that access the db t...
[16:39:40] <wikibugs>	 Analytics / Quarry: Add 'download as SQL' option - https://bugzilla.wikimedia.org/69191 (Yuvi Panda) NEW p:Unprio s:normal a:None So people can download the data, and put it in a local DB and run queries on the results on the query.  QUERYCEPTION.
[16:41:11] <wikibugs>	 Analytics / Quarry: Build a schema explorer - https://bugzilla.wikimedia.org/69192 (Yuvi Panda) NEW p:Unprio s:normal a:None Since a lot of people don't actually know how our data is structured and what / how they can query them as.
[16:41:22] <wikibugs>	 Analytics / Wikimetrics: session management - https://bugzilla.wikimedia.org/68833#c4 (nuria) Actually celery has worker signals: http://celery.readthedocs.org/en/latest/userguide/signals.html#worker-signals  I think we can tie to those to initiate and destroy sessions as those signals are tied to work...
[16:42:09] <wikibugs>	 Analytics / Quarry: Build a vagrant box for Quarry dev - https://bugzilla.wikimedia.org/69193 (Yuvi Panda) NEW p:Unprio s:normal a:None Reuse mw-vagrant, or use a new one.   Would need:  - ToolLabs account (for replica db access) - OAuth creds (self-serve)
[17:14:07] <wikibugs>	 Analytics / EventLogging: Helper for logging link clicks - https://bugzilla.wikimedia.org/52287#c8 (Matthew Flaschen) For this particular use case, there is also the ping attribute of the <a> element (https://developer.mozilla.org/en-US/docs/Web/HTML/Element/a#attr-ping), but I think that would require...
[17:26:22] <wikibugs>	 Analytics / EventLogging: Helper for logging link clicks - https://bugzilla.wikimedia.org/52287#c9 (nuria) The beacon API just made it into chrome, which, as you know, it is made for this very purpose.  https://dvcs.w3.org/hg/webperf/raw-file/tip/specs/Beacon/Overview.html
[17:55:36] <grrrit-wm>	 (CR) Yurik: [V: 2] Updated settings storage [analytics/zero-sms] - https://gerrit.wikimedia.org/r/150409 (owner: Yurik)
[18:00:18] <grrrit-wm>	 (PS5) Ottomata: [WIP] Notify Icinga about done webrequest datasets [analytics/refinery] - https://gerrit.wikimedia.org/r/152050 (owner: QChris)
[18:01:43] <grrrit-wm>	 (CR) Ottomata: "I'm tempted to make this workflow useable as a subworkflow from the actual check_sequence_stats workflow. That way we could still explici" [analytics/refinery] - https://gerrit.wikimedia.org/r/152050 (owner: QChris)
[18:14:44] <grrrit-wm>	 (CR) Dzahn: "do the sitemap links matter? <a href='http://en.wikipedia.org/wikistats/wikispecial/EN/Sitemap.htm" [analytics/wikistats] - https://gerrit.wikimedia.org/r/147876 (owner: Chmarkine)
[18:15:12] <grrrit-wm>	 (CR) Dzahn: [C: 2] use protocol relative url for image links on stats homepage [analytics/wikistats] - https://gerrit.wikimedia.org/r/147876 (owner: Chmarkine)
[18:15:28] <grrrit-wm>	 (Merged) jenkins-bot: use protocol relative url for image links on stats homepage [analytics/wikistats] - https://gerrit.wikimedia.org/r/147876 (owner: Chmarkine)
[18:15:45] <grrrit-wm>	 (CR) Dzahn: "cool, let's make a follow-up that removes the tabs and leading spaces" [analytics/wikistats] - https://gerrit.wikimedia.org/r/147876 (owner: Chmarkine)
[18:19:43] <grrrit-wm>	 (CR) Dzahn: "do we need to run something like puppet-merge here? not sure in analytics/wikistats about deploying" [analytics/wikistats] - https://gerrit.wikimedia.org/r/147876 (owner: Chmarkine)
[23:06:09] <leila>	 hi nuria, are you around?